INTERNATIONAL EARTH OBSERVATION INITIATIVES AND PROGRAMS: THE
CONTEXT FOR GLOBAL URBAN MAPPING
Martin Herold a and Christiane C. Schmullius a
a
ESA GOFC-GOLD Land Cover Project Office, Dep. of Geography, FSU Jena, Loebedergraben 32, 07743 Jena, Germany,
http://www.gofc-gold.uni-jena.de/, m.h@uni-jena.de.
KEY WORDS: global, urban, mapping, monitoring, GOFC-GOLD
ABSTRACT:
Besides the dominant focus of urban remote sensing research on local and regional scales, the URS2005 conference offers a special
session on “Global Observation of Urban Areas”. This session was jointly initiated by the GOFC-GOLD (Global observations of
forest and land cover) emphasizing that coordinated, consistent, and operational urban mapping and monitoring has to consider the
evolving efforts in international coordination and cooperation for land observations at global scales. Recent developments again
emphasize the need for sustained, harmonized, and validated earth observation products, i.e. in UNCED’s Agenda 21, UNFCCC and
the Kyoto protocol, the World Summit on Sustainable Development (WSSD) in Johannesburg 2002 and the related Group of Earth
Observations (GEO) formed in 2003 that has evolved to a Global Earth Observation System of Systems (GEOSS), the European
Global Monitoring for Environment and Security (GMES) initiative, as well as, the GCOS implementation plan calling on land cover.
These developments aim at long term goals but have to start and evolve from an international cooperation and consensus building
efforts, both in on the strategic level and in implementation activities. These developments will set the frame for global urban
observations with its main focus on continuity and consistency in earth observations that best fits user requirements.
1. INTRODUCTION
The persisting dynamic urban change processes, especially the
tremendous worldwide expansion of urban population and
urbanized area, affect and drive natural and human systems at all
geographic scales. Urbanization is the trigger for a variety of
other land change processes in natural and semi-natural
environments. Any operational efforts tailored at sustainable
and desirable future development have to consider urban
dynamics as one of the key human-induced processes for
understanding and managing our fast changing world.
Earth observation has been focused on mapping, monitoring and
understanding these urban phenomena for many years, however,
with more emphasize on local to regional scales. Global mapping
of human settlements faces particular challenging due to spatial
and spectral heterogeneity of urban environments, as well as,
their small and fragmented spatial configuration. In fact, there is
large disagreement between urban land represented in different
global land cover products (Fig.1, see also Schneider et al.,
2003). Different reasons can be mentioned: challenges in
mapping urban areas with coarse scale earth observation
systems, different definitions of “urban” and varying mapping
standards, issues with integrating different data sources,
problems in precise geo-location of spatial datasets, and the
importance of up -date information since urban areas are quickly
evolving. Thus, global urban mapping can, in most cases,
currently only marginally deliver what is needed to approach the
multitude of challenges resulting from continuous urbanization
and population growth.
The recent proliferation of new sources of data and tools for
data processing, analysis and modeling has provided and opened
up avenues for significant progress towards global observations
of urban patterns and dynamics. Due to the heterogeneity of
global urban characteristics, the key issue is to combine earth
observation indicators for characteristics and change in human
settlements. Sensors like MODIS or LANDSAT give spectral
evidence for built -up areas and the land cover configuration
within urban environments; night-time observations by DMSP
are a strong indicator of populated areas and population
distribution; SAR measurements emphasize the threedimensional characteristics of urban surfaces; thermal IR data
contain information about energy fluxes and local climatic
conditions.
São Paulo
Shanghai
Figure 1: Comparison and agreement of urban land in three
different global land cover products: IGBP DIS (urban areas
from Digital chart of the world), MODIS land cover (urban areas
from MODIS (2000), DMSP (1994/95) and ancillary data), and
GLC2000 (urban areas from DMSP (1994/95)).
Other key projects and programs have been looking at
representative urban agglomerations world wide to monitor
spatio-temporal urban dynamics. The overall goal of earth
observation is the mapping, monitoring, and analysis of urban
form and processes towards sup port and improvement of urban
modeling, management and planning efforts, and advances in
understanding urban phenomena. Sustainable development
activities benefit from the resulting better data, knowledge, and
information, if the int egrated framework includes a better
integration and communication of the earth observation results,
and works towards general acceptance of new and innovative
techniques in approaching urban dynamics.
In parallel to developments of data availability and scientific
research on their analysis, integration, and application, several
international earth observation initiatives are currently evolving.
The general goal is continuity and consistency in providing high
quality data in support of sustainable and desired future
development. GOFC-GOLD (Global observations of forest
cover and Global observations of land dynamics,
http://www.fao.org/gtos/gofc-gold) is technical panel of Global
Terrestrial Observing System (GTOS, http://www.fao.org/gtos)
and provides the interface between the strategic and the
implementation level. GOFC-GOLD provides a communication
and cooperation platform for actors involved in global earth
observation including data producers (e.g. space agencies, land
cover facilities), the scientific community, and data users (FAO,
UNEP, global modeling community etc.). As part the activities,
GOFC-GOLD jointly organized a special session on “Global
Observations of Urban Areas” at the URS2005 conference. The
overall goal of this special session is to provide a forum for
researchers to communicate and discuss issues and challenges in
coarse scale urban mapping and monitoring with respect to the
recent strategic developments in the global earth observation
arena. The aim of this paper is to provide an overview of current
strategic developments and their requirements, and a summary
of previous global urban mapping and monitoring activities and
related challenges to set the frame for the special session and
related discussions.
2. INTERNATIONAL LAND OBSERVATION
INITIATIVES AND PROGRAMS
This section describes the main strategic developments and
documents on global earth observations of land that have
become available or started with the last year. Previous
activities, e.g. the CEOS Global Urban Monitoring by the
Working Group on Information Systems and Services (WGISS),
are not described. Thus, all the described initiatives are very
active and emphasize the importance of earth observation as
basis for accurate and timely land surface data on global scales.
These current activities show that there many activities that
ultimately result further improvements in the global earth
observation arena.
2.1 Group on Earth Observation (GEO) and the Global
Earth Observation System of Systems (GEOSS)
Figure 2: The “big picture” in global earth observations:
organizations and agencies involved in defining requirements,
strategies and implementation activities.
The
Group
of
Earth
Observations
(GEO,
http://earthobservations.org/) was a direct follow up of the
World Summit on Sustainable Development held 2002 in
Johannesburg that called for strengthened cooperation and
coordination among global observing systems and research
programmes for integrated global observations. GEO emerged in
2003 from a consensus among governments and international
organizations that, while supporting and developing existing
Earth observation systems, more can and must be done to
strengthen global cooperation and Earth observations. The GEO
vision was formulated in the Washington Declaration adopted at
the Earth Observation Summit of 2003. Since this declaration in
2003, the GEO process has resulted in a variety of Earth
Observation Summits, the latest 6th Group on Earth
Observations meeting took place in Brussels, Belgium on 1415th February 2005.
The GEO process has outlined a framework document calling on
Global Earth Observation System of Systems (GEOSS). While
not legally binding, this document marks a crucial step in
developing the 10-Year Implementation Plan for the creation of
a comprehensive, coordinated, and sustained Earth observation
system or systems.
Table 1: GEOSS earth observation objectives for urban
mapping on global scales
Global earth observation
Societal benefit
requirements referring to urban
area
areas
Disasters
Human infrastructure
Health
Urban heat island and air quality
Population density
Land cover
Energy
Land use and land cover
Urban extent
Climate
Land cover
Water
Land use
Industrial water demand
Population density
Ecosystems
Population density
Agriculture
Land cover
Population density
Given the current draft of the GEOSS implementation plan
(Dec.2004) global urban mapping is mentioned in several
circumstances (Table 1). Several aspects of urban mapping are
described that several areas of societal benefit. It should be
noted that some of these observation requirements are benefitial
to several of these areas and are not mentioned twice. Urban
extent and land use/land cover maps are considered to be not yet
widely available or not yet adequately monitored globally but
could be within two to ten years. One parameter not mentioned
in Table 1 is transportation infrastructure that also is called on
in the GEOSS implementation plan.
Most of the urban mapping features (e.g. urban extent, land use
and land cover etc.) are considered to be not yet widely available
or not yet adequately monitored globally but could be within
two to ten years. The GEOSS plan also emphasizes the need for
integrative analysis of the earth observation mapping products,
i.e. gaps exist in the integration of relevant existing observation
systems, for example, integrating the global urban land
observations with data that characterizes the built environment,
chemical emission, and with indicators of environmental quality,
health and disease. Transportation infrastructure is also
mentioned as additional mapping objective with a strong link to
urban areas.
2.2 Integrated Global Observations of Land (IGOL)
The
Integrated
Global
Observing
Strategy
(IGOS
http://ioc.unesco.org/igospartners/) has been established in 1998.
Its main objective is the definition, development and
implementation of an Integrated Global Observing Strategy.
IGOS brings together efforts of a number of international
agencies concerned with global environmental issues, research,
and observing systems. IGOS theme documents are a primary
source of requirements for the development of GEOSS.
However, the IGOS-Partnership (IGOS-P) has not yet
considered many observational needs relating to many aspects
of the land, such as sustainable economic development, natural
resources management, conservation and biodiversity,
ecosystems (functioning, services), biogeochemical cycling,
multilateral
environmental
agreements
(development,
implementation), mandatory reporting and monitoring. A new
IGOS theme is currently being proposed on Integrated Global
Observations of the Land (IGOL). The main components of the
new theme are: land cover and land use, human settlement and
population, managed ecosystems, agriculture, pastoralism,
forestry, natural ecosystems, conservation, biodiversity,
sustainable use, soils, biogeochemical cycles, and elevation.
In terms of global urban mapping, IGOL emphasizes the need of
coordinated global urban observations with the fundamental
focus on reliable and spatially explicit settlement and population
databases, transportation infrastructure information, and
understanding urban change processes. Furthermore, there are
plans for an IGOS working group specifically tailored at socioeconomic data issues relevant to IGOS themes.
2.3 Global Monitoring for Environment and Security
(GMES )
The objective of this initiative is to "establish by 2008 a
European Capacity for Global Monitoring of Environment and
Security (GMES)" (http://www.gmes.info). GMES, as joint
initiative of the European Commission and the European Space
Agency (ESA), aims to support Europe's goals regarding
sustainable development and global governance, in support of
environmental and security policies, by facilitating and fostering
the timely provision of quality data, information, and
knowledge. As part of the GMES an “Urban Services ” section
(http://www.gmes-urbanservices.com/) has been established.
The objectives do not especially refer to global scale urban
mapping. GMES p riorities in urban mapping are set on local and
regional scales to assist urban management and security.
2.4 Millennium Ecosystem Assessment (MEA)
The
Millennium
Ecosystem
Assessment
(MEA ,
http://www.millenniumassessment.org/ ) is an UN initiated
international work program designed to meet the needs of
decision makers and the public for scientific information
concerning the consequences of ecosystem change for human
well-being and options for responding to those changes. Being
tailored at ecosystems, the observation requirements focus on
mixed patterns of human use and ecosystems which emphasizes
on the spatial extent of urban areas. A further focus is on
urbanization processes and population growth as driver of
ecosystem change on different spatial and temporal scales.
2.5 Global Climate
implementation plan
Observing
Systems
standardized mapping using the FAO Land Cover Classification
System (LCCS).
3. GLOBAL MAPPING AND MONITORING
ACTIVITIES
(GCOS)
In support of the United Nations Framework Convention on
Climate Change (UNFCCC), the Global Climate Observing
System (GCOS, http://www.wmo.ch/web/gcos/gcoshome.html)
has completed an implementation plan in October 2004 to
outline the requirements and actions to provide appropriate data
base in the implementation of the UNFCCC objectives and the
Kyoto protocols . The focus of this implementation plan is on
climate but specifically calls on land cover and changes (and use)
as important terrestrial variable to be derived from earth
observation.
2.6 GTOS-Coastal implementation plan
The GTOS implementation plan on coastal zones
(http://www.fao.org/gtos/doc/pub36.pdf)
emphasizes
the
importance of urban areas in this environment since about half
of the world population lives within 200 km from the coast. The
IGOS Coastal theme already puts a priority on urbanization in
coastal zones. Coastal GTOS implementation plan specifies
several aspects, i.e. the rate of change in population,
urbanization, and land use in coastal environments. The report
discusses “best available global datasets” and their current
limitations and prospects including the DOE Landscan Ambient
Population, the DMSP Night-time data, the global Landsat
mosaics, the ESA GLOBCOVER product and the MODIS land
cover/urban product and vegetation continuous fields datasets.
2.7 UN Global Land Cover Network (GLCN)
The
Global
Land
Cover
Network
(GLCN,
ftp://ftp.fao.org/docrep/fao/004/y3726e/y3726e00.pdf) launched
by FAO and UNEP, is an international coordinated effort with
the objective to provide direction, focus, guidance and standards
for harmonization of land cover mapping and monitoring at
national, regional and global levels. The initiative is based on the
recommendations of the Agenda 21 for coordinated, systematic
and harmonized collection and assessment of data on land cover
and environmental conditions. GLCN aims at generating
essential data needed for sustainable development,
environmental protection, food security and humanitarian
programmes of the United Nations, and of other international
and national institutions. The major objectives of this initiative
are on harmonization and standardization of classification of
cover types, the determination patterns of land-cover and its
associated change, projections of human response scenarios,
support to integrated global and regional modelling and the
global assessment of land cover for international conventions
and treaties. The GLCN strategic documents are currently in
development and will also address issues of urban mapping and
monitoring and related expectations for earth observations, and
Previous global activities have applied a variety of data sources,
mapping and monitoring strategies and analysis methods to
study urban phenomena on gl obal scales (Table 2). The most
comprehensive information on global urban dynamics has been
derived from statistical datasets describing demographic, socioeconomic and economic indicators of urban characteristics and
quality. On global scales, this information usually exists in
rather coarse spatial precision, with large time steps for
updating, and sometimes the data are not consistently available
for specific regions (e.g. developing countries). More detail can
be provided by earth observation. Remote sensing data sources
for coarse scale urban mapping have been multifaceted: optical
data (Schneider et al., 2003), thermal measurements (Hafner and
Kidder, 1999), active Radar data (Henderson and Xia, 1997,
Grey et al., 2003), and night-time-lights DMSP data (Imhoff et
al., 1997, Henderson et al. 2003). The data sources have been
used to study a variety of urban phenomena like urban
ecosystems (Miller and Small, 2003), urban climates (Voogt and
Oke, 2003), urban population (Sutton et al., 1997, Small, 2003,
Liu et al., 2005), health and disease (Tatem and Hay, 2004),
urban growth and change processes (Phinn et al., 2002, Seto and
Kaufmann, 2003, Herold et al., 2003) and others. So far,
comprehensive global urban monitoring programs have been
using selected rep resentative cities. These urban areas are
monitored in high spatial and temporal detail. The observations
and analysis provide general assumptions about ongoing
processes on coarser scale (Lavalle, et al., 2001, Beckel et al.,
2002).
4. CHALLENGES FOR GLOBAL URBAN
OBSERVATIONS
The initiatives presented in the previous sections emphasize
that basic global urban information is still lacking. T he best
strategy to overcome these limitations are coordinated
international mapping and monitoring efforts which include
earth observations as central data source. To be successful,
different challenges are mentioned in the strategic documents.
Despite a focus on global observations to data products have to
be derived, analyzed, and refined a multi-scale context. Urban
p henomena can be observed and show specific characteriscs on
different levels of spatial and temporal and detail. Linking coarse
scale (250-1000 m spatial resolution), fine-scale (20-50 m) and
very fine-scale (1-4 m) and in-situ observations should be
central perspective in any related activities. Continuity in earth
observations on all these scales are essential to support such
progress. Currently, coarse scale observations are widely
available (e.g. MODIS, MERIS, SPOT VEGETATION) and
will so in the future with systems like NPOESS. This level of
continuity does not exist for the other scales.
As much as spatial and temporal detail, the issue of thematic
and semantic accuracy will be of particular importance. In urban
mapping (and other fields) data products are often produced as
independent datasets with its individual semantic definitions.
Issues of harmonization, interoperability, and synergy between
different data sources and mapping products are important more
than ever. This includes the standardized development of
legends (e.g. using the FAO Land Cover Classification System)
and flexible thematic definitions (e.g. through continuous rather
than discrete mapping products, i.e. urban – rural gradient).
Also, ni teroperability and application in general assumes that
mapping products are independently validated.
community is invited to provide input and evaluate strategies to
further improve global urban observations. The themes for this
special session reflect the recent developments and
requirements:
§
§
§
§
§
§
Another essential requirement is the integration of the earth
observation mapping and monitoring products with existing
socio-economic/demographic information. These data are usually
non-spatial data or spatially aggregated. An integration
framework is presented in Figure 3. Earth observation measures
urban characteristics “bottom up” describing the outcome of
various processes at work (from structure to process). Socioeconomic drivers or specific urban models usually follow a
“top-down” approach by studying a pre-specified process of
urban change and the resulting spatio-temporal patterns (from
process to structure). Linking these two using spatial urban
theory, spatial analysis and modeling provides an appropriate
framework for monitoring, understanding, and modeling urban
phenomena. This information is essential to anticipate and
forecast future changes or trends of development, describe and
assess economic, ecological, and social impacts of urbanization.
Ultimately, an comprehensive integration framework will help
building the bridge between observation and use and help to
develop a common suite of urban indicators and information that
use useful to use and support exp lore different policies and
scenarios in support of planning and management decisions.
Figure 3. Conceptual framework for integration of earth
observation data and drivers and factors of urban processes.
5. FOCUS OF THE SPECIAL SESSION
With the special on “Global observations of urban areas” the
URS2005 conference, GOFC-GOLD is seeking input and
discussions to approach these challenges. The science
§
Tools and methods for mapping and monitoring urban areas
on global scales
Urban areas and global land cover products
Global trends of urban development from earth observation
Drivers/factors influencing global urban dynamics
Integration of earth observation data in coarse scale urban
modelling
Economic, ecological and social impacts of urban
development
Forecasts of future urban developments and prospects in
observing urban processes
Table 2: Overview of prominent urban mapping and monitoring initiatives and projects with focus on global/regional scales
The outcomes of this special session will hopefully results in a
synthesis describing prospects and scientific perspectives of
global urban observations that can used by organizations like
GOFC-GOLD to guide their future directions and efforts.
Henderson, M., E. T. Yeh, P. Gong, C. Elvidge and K. Baugh
2003. Validation of urban boundaries derived from global night time satellite imagery, International Journal of Remote Sensing,
24, 3, 595-609.
6. SUMMARY
Herold, M, Goldstein, N. C. & Clarke, K. C. 2003. The spatiotemporal form of urban growth: measurement, analysis and
modeling, Remote Sensing of the Environment, 86, 286-302.
Urban areas and their dynamics are one of the main drivers of
land change on local to global scales. With small and fragmented
spatial and spectrally heterogeneous characteristics, their
accurate mapping and monitoring has been challenged in the past
and certainly has not received as much attention as the global
observation of other land types such as forests. Urban mapping
requires different earth observation data sources and satellite
data analysis approaches than suitable for other land surface
types.
Different initiatives have provided interesting experiences and
explored different strategies for assessment of urban change on
global scales. The recent development in the strategic Earth
Observation arena specifies the requirements and expectations
for global urban monitoring programs. There is consensus that a
comprehensive and sustained global urban observatory has to be
built on coordinated international actions. Data products have to
be harmonized, accessible, validated, and flexible to provide a
better match between observations, data products, and user
requirements, i.e. as outlined in the UN strategic documents.
The establishment of a global urban observatory provides great
opportunities to establish successful cases for data integration
that is desired in many fields of global earth observation.
Synergy of satellite data from different sensors (DMSP,
MODIS, SAR etc.), linking observations to socio-economic and
demographic information, bridging across and among different
scales, and relating empirical measurements, spatial theory and
modeling have been proven to a successful in an urban context.
These potentials should be further elaborated on for global scale
assessments. Basically, the framework exists to improve global
urban observations. It is no up to earth observation community
to respond to the challenges.
Imhoff M.L., Lawrence W.T., Elvidge C.D., Paul T., Levine E.
and Privalsky M.V . 1997. Using nighttime DMSP/OLS images
of city lights to estimate the impact of urban land use on soil
resources in the United States, Remote Sensing of Environment,
59, 1, 105-117.
Lavalle C., Demicheli, L. Turchini, M., Casals -Carrasco, P., and
Niederhuber, M. 2001. Monitoring megacities: the
MURBANDY/MOLAND approach, Development in Practice,
11, 2-3, 350 – 357.
Liu, X. Clarke, K.C. and M. Herold 2005. Population Density
and Image Texture: A Comparison Study, Photogr ammetric
Engineering and Remote Sensing, in press.
Miller, R. B. and Small C. 2003. Cities from space: potential
applications of remote sensing in urban environmental research
and policy, Environmental Science and Policy, 6, 2, 129-137.
Phinn S., M. Stanford , P. Scarth , A. T. Murray , and P. T.
Shyy 2002. Monitoring the composition of urban environments
based on the vegetation-impervious surface-soil (VIS) model by
subpixel analysis techniques, International Journal of Remote
Sensing, 23, 10, 4131 - 4153
Schneider A, Friedl MA, Mciver DK, et al. 2003.
Mapping urban areas by fusing multiple sources of coarse
resolution remotely sensed data, Photogrammetric Engineering
and Remote Sensing, 69, 12, 1377-1386
Seto K.C. and Kaufmann R.K. 2003. Modeling the drivers of
urban land use change in the Pearl River Delta, China: Integrating
remote sensing with socioeconomic data. Land Economics, 79, 1,
106-121.
REFERENCES
Beckel, L., Kuehnen, A., Siebert, A., Lichtenegger, J., Gampe,
F., and Brisson P., 2002. MegaCities, ESA Bulletin-Europoean
Space Agency, 110, 81-87.
Small, C., 2003. Global Population Distribution and Urban Land
Use in Geophysical Parameter Space, Earth Interactions, 8, 8, 118.
Grey, W. M. F., Luckman, A. J, and Holland D. 2003. Mapping
urban change in the UK using satellite radar interferometry,
Remote Sensing of Environment, 87, 1, 16-22.
Tatem, A. J. and Hay S. I. 2004. Measuring urbanization
pattern and extent for malaria research: A review of remote
sensing approaches, Journal of Urban Health Bulletin of the
New York Academy of Medicine. 81, 3, 363-376.
Hafner J. and Kidder S.Q. 1999. Urban heat island modeling in
conjunction with satellite-derived surface/soil parameters,
Journal of applied Meteorology, 38, 4, 448-465.
Voogt J,A, and Oke T.R. Thermal remote sensing of urban
climates Remote Sensing of Environment, 86, 3, 370-384.
Henderson F.M. and Xia Z.G. 1997. SAR applications in
human settlement detection, population estimation and urban
land use pattern analysis: A status report, IEEE Transactions on
Geoscience and Remote Sensing, 35, 1, 79-85.